Safe Sequential Path Planning Under Disturbances and Imperfect Information

TL;DR

This paper enhances sequential path planning for multi-UAV systems by incorporating disturbances and imperfect information, making safety guarantees more practical in real-world scenarios.

Contribution

It introduces three novel methods to account for disturbances and imperfect knowledge within the sequential path planning framework, expanding its applicability.

Findings

Methods successfully handle disturbances in simulations.

Approaches accommodate varying levels of information sharing.

Enhanced safety guarantees in multi-UAV path planning.

Abstract

Multi-UAV systems are safety-critical, and guarantees must be made to ensure no unsafe configurations occur. Hamilton-Jacobi (HJ) reachability is ideal for analyzing such safety-critical systems; however, its direct application is limited to small-scale systems of no more than two vehicles due to an exponentially-scaling computational complexity. Previously, the sequential path planning (SPP) method, which assigns strict priorities to vehicles, was proposed; SPP allows multi-vehicle path planning to be done with a linearly-scaling computational complexity. However, the previous formulation assumed that there are no disturbances, and that every vehicle has perfect knowledge of higher-priority vehicles' positions. In this paper, we make SPP more practical by providing three different methods to account for disturbances in dynamics and imperfect knowledge of higher-priority vehicles' states. Each method has different assumptions about information sharing. We demonstrate our proposed methods in simulations.